High Energy Physics and Grid

1. High Energy Physics and Grid Oxana SmirnovaFYS225/FKF050February 27, 2007, Lund

2. High Energy Physics: quick intro • High Energy Physics studies elementary particles and forces that constitute the matter • Also known as Particle Physics • Is the most fundamental of all the physical sciences • The ultimate goal: to understand the origin of the Universe • We’ve learned a lot over the past years, yet more is unknown www.nordugrid.org

3. Leptons Strong Electromagnetic Electric Charge Gluons (8) Photon Tau Tau -1 0 Neutrino Muon Muon -1 0 Neutrino Quarks Atoms Light Electron Electron -1 0 Neutrino Chemistry Mesons Electronics Baryons Nuclei Quarks Weak Gravitational Electric Charge Bosons Graviton ? Bottom Top -1/3 2/3 (W,Z) Strange Charm -1/3 2/3 Neutron decay Down Solar system Up Beta radioactivity -1/3 2/3 Galaxies Neutrino interactions Black holes Burning of the sun each quark: R , B , G 3 colours The particle drawings are simple artistic representations Standard Model of forces and particles www.nordugrid.org

4. From Big Bang to now: the theory www.nordugrid.org

5. Still, we know too little about ourselves! • Can gravity be included in a theory with the other three interactions? • Why do the particles have the masses we observe, and what is the origin of mass? • How many space-time dimensions do we live in? • Are the known elementary particles fundamental or do they possess structure? • Why is the electrical charge on the electron equal and opposite to that on the proton? • Why are there three generations of quarks and leptons? • Why is there overwhelmingly more matter than anti-matter in the Universe? • Are protons unstable? • What is the nature of the dark matter that pervades our Galaxy? • Are there new states of matter at exceedingly high density and temperature? • Do the neutrinos have mass, and if so why are they so light? www.nordugrid.org

6. Large Hadron Collider:World’s biggest accelerator at CERN Will start operating in fall 2007 www.nordugrid.org

7. Collisions at LHC • Will help to solve: • The mass problem – discover Higgs particle? • The Dark Matter of the Universe – discover supersymmetric particles? • Matter vs Antimatter – CP violation? www.nordugrid.org

8. ATLAS detector at LHC www.nordugrid.org

9. Event generation (Pythia) Detector simulation (Geant4) 100011110101110101100101110110100 Hit digitization Reconstruction Analysis data preparation Analysis, results (ROOT) “Full chain” of HEP data processing www.nordugrid.org Slide adapted from Ch.Collins-Tooth and J.R.Catmore

10. Monte Carlo data production flow (10 Mevents) • Very different tasks/algorithms (ATLAS experiment in this example) • Single job lasts from 10 minutes to 20 hours • Most tasks require large amounts of input and produce large output data www.nordugrid.org

11. Software for HEP experiments • Massive pieces of software • Written by very many different authors in different languages (C++, Java, Python, Fortran) • Dozens of external components • Occupy as much as ~10 GB of disk space each release • Frequent releases • Every experiment can release as often as once a month during the preparation phase (which is now for LHC) • Difficult to set up outside the lab • Experiments can not afford supporting different operating systems and different computer configurations • For a small university group it is very difficult to manage different software sets and maintain hardware • ALICE, ATLAS, H1, PHENIX etc – all in many versions • Solution: use external computing resources www.nordugrid.org

12. HEP computing specifics • Data-intensive tasks • Large datasets, large files • Lengthy processing times • Large memory consumption • High throughput is necessary • Very distributed resources • Distributed computing resources of modest size • Produced and processed data are hence distributed, too • Issues of coordination, synchronization and authorization are outstanding • HEP is by no means unique in its demands, but we are first, we are many, and we badly need it www.nordugrid.org

13. The Grid Supercomputer Sewage Water Electricity Radio/TV PC Farm Workstation Internet Grid Grid to the rescue • Distributed supercomputer, based on commodity PCs and fast WAN • Access to the great variety of resources by a single pass – certificate • A possibility to manage distributed data in a synchronous manner • A new commodity www.nordugrid.org

14. From distributed resources … • Present situation: • cross-national projects • users and resources in different domains • separate access to each resource www.nordugrid.org

15. … to World Wide Grid • Future: • multinational projects • resources location is irrelevant • “plug-n-play” access to all the resources www.nordugrid.org

16. Origin of the word “Grid” • Coined by Ian Foster and Carl Kesselman around 1997 • Refers to computing grids as analogy of power grids • Many producers • Competing providers • Simple for end-users • Spelled “grid” or “Grid” • Except in French: “Grille de calcul” www.nordugrid.org

17. Grid as a result of IT progress • Network vs. computer performance: • Computer speed doubles every 18 months • Network speed doubles every 9 months • 1986 to 2000: • Computers: 500 times faster • Networks: 340000 times faster • 2001 to 2010 (projected): • Computers: 60 times faster • Networks: 4000 times faster Bottom line: CPUs are fast enough; wide area networks are very fast – gotta make use of it! www.nordugrid.org Slide adapted from the Globus Alliance

18. How did Grid appear in Lund • Back in 2001, High Energy Physics Institutes from Scandinavia wanted to share their computing resources and jointly contribute to CERN/LHC computing • We needed Grid • The Grid hype just begun… • … and we created a NorduGridproject (Lund, Uppsala, Copenhagen,Oslo, Helsinki and many others) • No production ready grid software (middleware) was available or seenon the horizon in fall 2001 • In February 2002, NorduGrid boldlydecided to develop own Gridmiddleware • Was baptized ARC, for Advanced Resource Connector • Since May 2002 ARC is extensively used in ATLAS production and other scientific computing projects www.nordugrid.org

19. ARC in a nutshell Goal: no single point of failure www.nordugrid.org

20. ARC is not the only one… www.nordugrid.org Slide adapted from Vicky White, FNAL

21. Grids in ATLAS • All ATLAS production today is done via Grid • There are 20+ Grid flavors out there • Almost all are incompatible with each other • Almost all are tailored for a specific application and/or specific hardware • ATLAS makes use of only 3 Grid flavors: • gLite – formerly known as LCG • ARC – formerly known as NorduGrid • OSG – formerly known as Grid3 • They are still incompatible with each other • A lot of interoperability efforts take place www.nordugrid.org

22. ARC and ATLAS • Via NorduGrid/ARC, Nordic countries contribute to ATLAS Data Challenges since 2002 • Resources donated to ATLAS by national Grid projects, enthusiastic owners • Highly heterogeneous (OS: Fedora Core N, Red Hat, Debian, Gentoo,...; LRMS: PBS/Torque, Condor, SGE…) • No common policies enforced • Loosely coupled • Currently, ca 10% of ATLAS production tasks • Only 2 persons in charge of the production • Highest resource usage efficiency, reliability • Accumulated ~40TB of ATLAS data in ~50 locations • Includes e.g. Ljubljana – still, indistinguishable for jobs and outside users www.nordugrid.org

23. ATLAS Monte Carlo production with NorduGrid/ARC www.nordugrid.org

24. Next step: Nordic Tier1 • WLCG: Worldwide LHC Computing Grid • A CERN project aiming to provide HEP computing infrastructure • Tiered structure: Tier0 at CERN, a dozen of regional Tier1s, many local Tier2s etc • WLCG Tier1 is primarily a set of services: • 24/7 on-call support system • Infrastructure: network, power, cooling, safety etc • Authorization, specific software for entire multinational VOs • Job submission interface • Data indexing service • Storage resource management interface • File transfer services between Tiers • Experiment-specific interfaces (“VOBoxes”) • Database service (ORACLE) • Other: information system, monitoring, logging etc • Resource commitments (processing power, storage, bandwidth) are expressed in WLCG MoU www.nordugrid.org

25. Where do I start on my way to Grid? • Ask local sysadmin to install a Grid client • NorduGrid standalone client is the easiest: you can install it yourself • Apply for a “passport”: the Grid certificate (grid-cert-request) • NorduGrid Certification Authority is appropriate for SU employees • Ask a knowledgeable person which Grid is adopted by your collaboration/group • Apply for a “visa”: become an appropriate Virtual Organization (VO) member • Read the manual Rule of thumb: do not keep your credentials on a public computer. USB memory key is a good choice. www.nordugrid.org

26. Conclusion • High Energy Physics is the major consumer of Grid technologies • Every HEP researcher sooner or later will have to learn grid basics • HEP community invests massive efforts into grid development • If grid won’t help, it is unclear what would be the “backup solution” • The data will eventually be processed, the question is – how soon and how accurate • Many other sciences are on-looking • Bioinformatics and radioastronomy appear to be the next in line • Huge data volumes, trivially parallel processing, distributed user base • See you in the Grid-space! www.nordugrid.org